Quantum interference in the control of molecular processes

P. Brumer,M. Shapiro

DOI: https://doi.org/10.1007/978-94-011-2642-7_22

1997-12-15

Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences

Abstract:

What problem does this paper attempt to address?

Measure liquid surface tension by fitting lying droplet profile

Hao Tang

DOI: https://doi.org/10.1016/j.measurement.2021.110379

2022-12-08

Abstract:This study developed a simple method to measure liquid surface tension based on curve fitting technique. In this study, the explicit lying droplet profile function in terms of surface tension was derived by solving 2-D Young-Laplace equation exactly, and the derived profile function was superior to the existing approximate profile functions. By fitting the derived profile function with the detected droplet profile image data, the liquid surface tension was then obtained. The proposed method does not rely on any new instrument, it is quite different in principle and simpler than the traditional methods.

Fluid Dynamics
Droplet deformation in steady fluidic flows enables robust, accessible, high-throughput surface tension measurements

Sabrina Marnoto,Evyatar Shaulsky,Avi J. Patel,Sara M. Hashmi

DOI: https://doi.org/10.26434/chemrxiv-2024-p8fz6

2024-09-17

Abstract:Impressive improvements in the ability of microfluidic devices to reliably fabricate a wide variety of droplet, capsule, and particle architectures necessitate comparable improvements in techniques to measure and characterize these materials in line. Measurement and control of droplet surface tension are needed to ensure droplet stability while minimizing excess use of surfactants. Standard techniques to measure surface tension typically measure one droplet at a time. Fluidics can be leveraged to measure surface tension in-line with droplet formation. Typically, this requires channel geometries that enable extensional flow. Transient relaxation of deformed droplets is then measured at the exit of a constriction. We propose an alternative approach, in which a single value of steady deformation within a constriction is used to measure surface tension. We flow aqueous droplets in mineral oil through a series of increasingly narrow constrictions and measure steady deformation. In a subset of experiments, droplets contain varying concentrations of polyethylene glycol diacrylate, a common blank-slate hydrogel polymer. We calculate surface tension using Taylor’s small deformation theory, which describes the relationship between steady deformation in shear flows and the Capillary number, the ratio of the applied viscous stress to restoring surface tension stress. We measure surface tension over a wide range of surfactant concentration. Validation using both the transient deformation fluidic approach and pendant droplet measurements demonstrates the viability of our approach. Importantly, our results suggest that steady state measurements of deformation in pressure driven flows provide accurate assessments of surface tension, even when droplets are slightly confined. The use of multiple constrictions allows measurement of hundreds of droplets at several distinct shear rates without the need to vary control parameters. This steady deformation approach represents a readily-accessible option for measuring surface tension of micro-scale droplets in pressure driven flow through rectangular channels.

Chemistry
Droplet Evolution Prediction in Material Jetting Via Tensor Time Series Analysis

Luis Javier Segura,Zebin Li,Chi Zhou,Hongyue Sun

DOI: https://doi.org/10.1016/j.addma.2023.103461

IF: 11

2023-01-01

Additive Manufacturing

Abstract:Material Jetting (MJ) process is an additive manufacturing process that is able to produce structures with high resolutions. The performance and quality of the MJ printed parts extensively rely on the droplet morphology and behavior. However, obtaining consistent and stable droplet morphology and behavior is difficult because the droplets are very sensitive to different material and process parameters. Testing/studying all these parameter combinations is time-consuming due to the high-experimental and high-computational costs. We thus study the prediction of droplet behaviors under different material and process parameters. To achieve this, we propose to leverage the underlying relationships shared across droplet evolution behaviors with diverse material and process parameters (referred to as “cross-linked” relationship hereafter) as well as the spatial–temporal relationships of droplet evolution, and capture these with the Network of Tensor Time Series (NET3). The distinct droplet behaviors are regarded as co-evolving time series (i.e., Tensor Time Series (TTS)) since they share the same physics principles. In particular, we capture the cross-linked and spatial relationships of TTS by the Tensor Graph Convolutional Network (TGCN), and capture the temporal relationship of TTS by the Tensor Recurrent Neural Network (TRNN), respectively. The features from TGCN and TRNN are passed to Multilayer Perceptron (MLP) for predicting future droplet behaviors. The proposed methodology is demonstrated with simulated (i.e., physics-based models) and experimental (i.e., MJ printing observation with vision system) droplet evolution videos and is able to accurately and efficiently make predictions for seen (i.e., forecasting for a future time) and unseen (i.e., new droplet evolution sequence prediction) material/process parameters.
Pinning Effects of Wettability Contrast on Pendant Drops on Chemically Patterned Surfaces

Liang Hu,Yao Huang,Wenyu Chen,Xin Fu,Haibo Xie

DOI: https://doi.org/10.1021/acs.langmuir.6b03318

IF: 3.9

2016-01-01

Langmuir

Abstract:The morphology and dynamics of the pendant drops attached to chemically patterned surfaces (pattern-pinned pendant drops) with different hydrophilic/hydrophobic contrasts were investigated experimentally and numerically. During the experiments, the evolution of the contact angle and the maximum drop volume were found to be different from those of traditional pendant drops, whose contact line is pinned on the edge of the tips (tip-pinned pendant drops), and the deviation is related to both the pattern radius and the wettability contrast. Then, a hypothesis was proposed to illustrate the behavior of the contact line after it reached the pattern boundary, based on the premise that the pattern boundary possessed a certain width or fuzziness. It was concluded that the special phenomena in this case were due to the movement of the contact line, and the maximum contact radius was presented as a key parameter for the pattern-pinned drops, which is directly related to the stability and the maximum volume of the drops. Furthermore, through a simulation study on pattern pinned pendant drops, the vibration performance of the meniscus was revealed as a superposition of two vibration behaviors including a low-frequency vibration due to the inertia effects and a high-frequency vibration due to the surface tension gradient within the boundary region. In addition, the hypothesis proposed above was also verified. Finally, a forecasting model to predict the maximum contact radius for the pattern-pinned pendant drops was built for different liquids and pattern wettabilities. This allows us to effectively design and optimize chemically patterned surfaces to achieve a desired pinning function or a pendant drop with desired properties.
Unsupervised Learning for the Droplet Evolution Prediction and Process Dynamics Understanding in Inkjet Printing

Jida Huang,Luis Javier Segura,Tianjiao Wang,Guanglei Zhao,Hongyue Sun,Chi Zhou

DOI: https://doi.org/10.1016/j.addma.2020.101197

IF: 11

2020-01-01

Additive Manufacturing

Abstract:Droplet jetting behavior largely determines the final drop deposition quality in the inkjet printing process. Forming such behavior is governed by the fluid flow pattern. Therefore, a measurement of the flow pattern is of great importance for improving the printing quality of the inkjet printing process. Most of the current works use static images for the study of the drop evolution process. The problem of the static images is that the images cannot recognize the motion information (i.e., temporal transformation) of the droplet. Thus the information of the jetting process in the temporal domain will be lost. Instead of using the images, this paper takes the video data as the study subject to investigate the droplet evolution behavior in the inkjet printing process. Moreover, this paper introduces a deep learning method for the study of such video data. Compared to most of the current learning approaches conducted in a supervised/semi-supervised manner for manufacturing process data, we propose an unsupervised learning method for studying the flow pattern of the droplet, which does not require well-defined ground-truth labels. Regarding the spatial and temporal transformation of the droplet in video data, we apply a deep recurrent neural network (DRNN) to implement the proposed unsupervised learning. To verify the hypothesis that the proposed method can learn a latent representation for reproducing original data, the proposed DRNN is trained and tested on both simulation and experimental datasets. Experimental results demonstrate that the proposed method can learn latent representations of the droplet jetting process video data, which is very useful for the prediction of the droplet behavior. Furthermore, through latent space decoding, the learned representations can infer the droplet forming stimulus parameters such as material properties, which would be very helpful for further understanding of the process dynamics and achieving real-time in-situ droplet deposition quality monitoring and control.
A deep learning perspective on electro-hydrodynamic micro-droplet interface deformation characteristics

Guanqiu Hao,Enhui Li,Jia-Xin Li,He Zhang,Mingliang Zhong,Xiangdong Liu,Ji-Xiang Wang,Yongping Chen

DOI: https://doi.org/10.1016/j.ces.2023.118772

IF: 4.7

2023-04-25

Chemical Engineering Science

Abstract:Micro-droplet electro-hydrodynamic deformations under combined shear flow and electric fields were investigated experimentally where droplet deformation mechanisms were revealed. Here two deep learning algorithms (DNN and DeepFM) were applied for better prediction accuracy in deformation characteristics (deformation parameter D and tilt angle φ). The DNN model showed better universal predicting accuracy (verified not for our experimental data but for data from previous publications). Specifically, the majority of DNN-based predictions were within a relative error of ±8%, gaining a major superiority compared with previous prediction modeling. Additionally, the use of a unique smart-sized training dataset for high accuracy without a large-scale data pool is another innovation, pioneering the deep-learning-based high-accuracy small-sample modeling. Considering the low accuracy from traditional modeling and the reduction in data scale requirement of the deep-learning-based method developed here, such a method can be extended to other complex multiphysics-controlled microfluidic droplet modeling in chemical engineering.

engineering, chemical
Accurate and robust static hydrophobic contact angle measurements using machine learning

Daniel Shaw,Ran Liang,Tian Zheng,Jianzhong Qi,Joseph Berry

DOI: https://doi.org/10.26434/chemrxiv-2024-6nhr1

2024-03-26

Abstract:Recent years has seen a surge in new approaches to material discovery. However, in order to validate and characterise these novel materials that are being discovered at unprecedented speeds, methods of experimental characterisation must match the pace of discovery. This is only possible through the use of high throughput autonomous approaches. % Automation addresses sources of human error To address this need, and to overcome the shortcomings in present methods, we present a machine learning (ML) approach to static contact angle measurement that is more accurate and faster than current best practice methods. The ML model was trained on a large data-set (> 7.2 million) generated via solutions of the Young-Laplace equation where the contact angle is known a priori, removing all sources of error from human input. The data-set included the effects of surface roughness, gravity, the size of drop relative to the image, reflections of the drop on the surface, and contact angles from 110° to 180°. The presented ML model (valid for contact angles >110°), in combination with a new automated image and contour processing approach, is shown to be more accurate than other methods when benchmarked against an experimental data-set, with an estimated error of 1 degree. The ML model is also two orders of magnitude faster at predicting contact angles than Young-Laplace fitting (the current best practice approach). The accuracy and speed of the presented approach provides a viable pathway towards robust and reproducible high-throughput contact angle analysis. The open-source software Conan-ML is provided for use and development of new approaches to goniometry.

Chemistry
Machine learning (ML)-assisted surface tension and oscillation-induced elastic modulus studies of oxide-coated liquid metal (LM) alloys

Kazi Zihan Hossain,Sharif Amit Kamran,Alireza Tavakkoli,M Rashed Khan

DOI: https://doi.org/10.1088/2515-7639/acf78c

2023-09-26

Journal of Physics: Materials

Abstract:Abstract Pendant drops of oxide-coated high-surface tension fluids frequently produce perturbed shapes that impede interfacial studies. Eutectic gallium indium or Galinstan are high-surface tension fluids coated with a ∼5 nm gallium oxide (Ga 2 O 3 ) film and falls under this fluid classification, also known as liquid metals (LMs). The recent emergence of LM-based applications often cannot proceed without analyzing interfacial energetics in different environments. While numerous techniques are available in the literature for interfacial studies- pendant droplet-based analyses are the simplest. However, the perturbed shape of the pendant drops due to the presence of surface oxide has been ignored frequently as a source of error. Also, exploratory investigations of surface oxide leveraging oscillatory pendant droplets have remained untapped. We address both challenges and present two contributing novelties- (a) by utilizing the machine learning (ML) technique, we predict the approximate surface tension value of perturbed pendant droplets, (ii) by leveraging the oscillation-induced bubble tensiometry method, we study the dynamic elastic modulus of the oxide-coated LM droplets. We have created our dataset from LM’s pendant drop shape parameters and trained different models for comparison. We have achieved >99% accuracy with all models and added versatility to work with other fluids. The best-performing model was leveraged further to predict the approximate values of the nonaxisymmetric LM droplets. Then, we analyzed LM’s elastic and viscous moduli in air, harnessing oscillation-induced pendant droplets, which provides complementary opportunities for interfacial studies alternative to expensive rheometers. We believe it will enable more fundamental studies of the oxide layer on LM, leveraging both symmetric and perturbed droplets. Our study broadens the materials science horizon, where researchers from ML and artificial intelligence domains can work synergistically to solve more complex problems related to surface science, interfacial studies, and other studies relevant to LM-based systems.
Design and implementation of an experimental device by the pendant drop method for measuring surface tension of pulmonary fluid

Zhenglong Chen,Dan LV,Chunyuan Zhang,Yuanyang Liu,Jingxin Sha,Linhong Deng,Ming Zhong,Yuanlin Song

DOI: https://doi.org/10.21203/rs.3.rs-1938535/v1

2022-01-01

Abstract:Abstract The quantitative measurement of pulmonary fluid surface tension in many lung diseases including respiratory distress syndrome (RDS), pneumonia, COVID-19 and acute respiratory distress syndrome (ARDS) has great physiological and clinical significance. The conventional techniques like film balances and bubble methods, however, are hard to use in the clinical practice due primarily to a limitation of small sample volume. In this work, we developed an experimental device based on the pendant drop method to clinically measure surface tension of pulmonary fluids in ARDS patients. We provided a detailed discussion of the theoretical background to the design of hardware and software as well as validation of measurement precision. The measured surface tensions of the two test liquids, pure water and dimethyl silicone oil, are 72.35±1.00 mN⁄m and 21.38±0.61 mN⁄m, respectively, in good agreement with the corresponding literature values of 72.02 mN⁄m and 21.10 mN⁄m at room temperature of 25℃.Using this experimental setup, we tentatively measured surface tensions of four pulmonary fluid samples from several mechanically ventilated ARDS patients. The results showed that the measured surface tension at 25±1℃ is within the range of 23.8～53.9 mN/m, comparable with reported values in the literature. Further, the calculation of the Worthington number showed that the data obtained are sufficiently accurate. We came to the conclusion that the pendant drop tensiometry is a powerful tool in the studies of pulmonary fluid surface properties due to a small sample size, high degree automation and freedom from contamination.
Full-profile Fit Pendent Drop Method for Surface Tension Measurements

ZHOU Bin,LI Siwei,CHEN Zhiyong,ZHANG Rong

DOI: https://doi.org/10.16511/j.cnki.qhdxxb.2016.25.047

2016-01-01

Abstract:The accuracy and stability of surface tension measurements are improved fitting the entire surface profile in the pendent drop method with an algorithm to determine the initial value for two-dimensional optimization.Moreover,the error function was modified by a weighting factor which reduced the influence of the droplet profile deviation point on the surface tension measurement.The full-profile fitting method was used to calculate the surface tension of water and ethanol at 25℃ and atmosphere.The final results show the standard deviation was reduced by 40％ compared with the previous method,which verifies the stability and repeatability of this full-profile fitting pendent drop method.
Axisymmetric Drop Shape Analysis: Computational Methods for the Measurement of Interfacial Properties from the Shape and Dimensions of Pendant and Sessile Drops.

A. W. Neumann,O. Rio

DOI: https://doi.org/10.1006/JCIS.1997.5214

IF: 9.9

1997-12-15

Journal of Colloid and Interface Science

Abstract:State-of-the-art axisymmetric drop shape analysis (ADSA) techniques for the computation of interfacial tensions and contact angles by fitting the Laplace equation of capillarity to the shape and dimensions of pendant and sessile drops are presented. More accurate, efficient, and reliable versions of the technique for the measurement of contact angles from the volume and diameter of sessile drops [axisymmetric drop shape analysis-diameter (ADSA-D)] and for interfacial tension measurements from a series of arbitrary profile coordinates of sessile and pendant drops [axisymmetric drop shape analysis-profile (ADSA-P)] have been developed. Advanced numerical methods have been used to improve the numerical stability and global convergence, for more accurate results and a wider range of applicability of the methods. A new technique called axisymmetric drop shape analysis-height and diameter (ADSA-HD) has been developed to estimate interfacial tensions from the height and diameter of sessile and pendant drops. Numerical simulations using numerically generated drop profiles were used to evaluate the accuracy and applicability of the methods. Copyright 1997 Academic Press.

Chemistry,Medicine,Materials Science,Engineering
Deep Learning Method for Determining the Surface Elastic Moduli of Microstructured Solids

Sang Ye,Wei-Zhi Huang,Min Li,Xi-Qiao Feng

DOI: https://doi.org/10.1016/j.eml.2021.101226

IF: 4.728

2021-01-01

Extreme Mechanics Letters

Abstract:Surface microstructures greatly affect many physical and mechanical properties of solid materials. Surface constitutive relations are often required to analyze the mechanical responses of materials and structures at the micro scales, but it remains a challenge to determine the surface constitutive parameters of microstructured materials. In this paper, a machine learning-based approach is proposed to predict the surface elastic properties of materials with complex microstructures on their surfaces. Using a data set generated from the finite element method, we demonstrate that the trained deep neural network can efficiently provide an accurate mapping between the equivalent surface elastic properties (e.g., elastic moduli and Poisson’s ratios) and the mechanical and geometric features of surface microstructures. This study provides an efficient way to evaluate the surface properties of materials and promises for wide applications in the analysis of micro-/nanostructured materials and devices.
Exploring droplet oscillation dynamics in surface tension measurements

Kiana Fahimi,Lutz Mädler,Nils Ellendt

DOI: https://doi.org/10.1007/s00348-024-03926-6

IF: 2.797

2024-11-29

Experiments in Fluids

Abstract:This study builds upon prior research by exploring droplet oscillation dynamics for surface tension determination using a drop-on-demand high-temperature droplet generator. Computational fluid dynamics (CFD) simulations were conducted to analyse frequency shifts over time, comparing two different materials with consistent results. The findings suggest potential for developing correction factors for oscillations with larger initial deformations. Additionally, frequency shifts relative to evolving aspect ratios of droplets starting with higher initial deformations were compared. Corrective measures can be applied, particularly beneficial for short-term measurements based on image analysis with minimal overall frequency shift. Slight asymmetry in oscillation with increasing aspect ratio could be accredited to droplet cross-sectional geometry or energy availability for returning prolate droplets to a spherical state. Experimental results indicated minimal frequency shift within a measurement duration of up to 40 ms, affirming the adequacy of using a fitted sine function without a time-dependent frequency term for overall frequency determination. A dimensionless criterion can be used to filter out unsuitable droplets. A temperature-dependent surface tension trend for AlCu10 alloy consistent with literature findings is introduced.

engineering, mechanical,mechanics
Surface tensiometry of phase separated protein and polymer droplets by the sessile drop method

Mahdiye Ijavi,Robert W. Style,Leonidas Emmanouilidis,Anil Kumar,Sandro M. Meier,Alexandre L. Torzynski,Frédéric H. T. Allain,Yves Barral,Michel O. Steinmetz,Eric R. Dufresne

DOI: https://doi.org/10.48550/arXiv.2007.13502

2020-07-27

Abstract:Phase separated macromolecules play essential roles in many biological and synthetic systems. Physical characterization of these systems can be challenging because of limited sample volumes, particularly for phase-separated proteins. Here, we demonstrate that a classic method for measuring the surface tension of liquid droplets, based on the analysis of the shape of a sessile droplet, can be effectively scaled down for this application. The connection between droplet shape and surface tension relies on the density difference between the droplet and its surroundings. This can be determined with small sample volumes in the same setup by measuring the droplet sedimentation velocity. An interactive MATLAB script for extracting the capillary length from a droplet image is included in the ESI.

Soft Condensed Matter
Machine learning materials physics: Deep neural networks trained on elastic free energy data from martensitic microstructures predict homogenized stress fields with high accuracy

K. Sagiyama,K. Garikipati

2019-01-02

Abstract:We present an approach to numerical homogenization of the elastic response of microstructures. Our work uses deep neural network representations trained on data obtained from direct numerical simulation (DNS) of martensitic phase transformations. The microscopic model leading to the microstructures is based on non-convex free energy density functions that give rise to martensitic variants, and must be extended to gradient theories of elasticity at finite strain. These strain gradients introduce interfacial energies as well as coercify the model, enabling the admission of a large number of solutions, each having finely laminated microstructures. The numerical stiffness of these DNS solutions and the fine scales of response make the data expensive to obtain, while also motivating the search for homogenized representations of their response for the purpose of engineering design. The high-dimensionality of the problem is reduced by training deep neural networks (DNNs) on the effective response by using the scalar free energy density data. The novelty in our approach is that the trained DNNs also return high-fidelity representations of derivative data, specifically the stresses. This allows the recapitulation of the classic hyperelastic response of continuum elasticity via the DNN representation. Also included are detailed optimization studies over hyperparameters, and convergence with size of datasets.

Computer Science,Physics,Materials Science
Deep Learning Regressors of Surface Properties from Atomic Force Microscopy Nanoindentations

João Ferreira,Luís R. L. Pacheco,Marco P. L. Parente

DOI: https://doi.org/10.3390/app14062376

2024-03-12

Applied Sciences

Abstract:Atomic force microscopy (AFM) is a powerful technique to study the nanomechanical properties of a wide range of materials at the piconewton level. AFM force–indentation curves can be fitted with appropriate contact models, enabling the determination of material properties for a given sample. However, the analysis of large datasets comprising thousands of curves using conventional methods presents a time-intensive challenge. As a result, there is an increasing interest in exploring alternative methodologies, such as integrating machine learning (ML) models to streamline and improve the efficiency of this process. In this work, two data-driven regressors were tuned to predict the Young’s modulus and adhesion energy from force–indentation curves of soft samples (Young’s modulus up to 10 kPa). Both models were trained exclusively on synthetic data derived from the contact theories developed by Hertz as well as Johnson, Kendall and Roberts (JKR). The PyTorch library was employed to build and train the models; then, the key hyperparameters were refined by implementing the optimization framework Optuna. The first model was successfully tested with synthetic and experimental curves from AFM nanoindentations, and the second presented promising results on the synthetic data. Our work suggests that experimental data may not be essential for training data-driven models to predict surface properties from AFM nanoindentations. By delivering accurate predictions in a computationally efficient way, our regressors validate the potential of a deep learning approach in exploring AFM nanoindentations and motivate further development of similar strategies to overcome current limitations in AFM postprocessing.

Computer Science,Engineering,Materials Science
Data-driven low-dimensional model of a sedimenting flexible fiber

Andrew J Fox,Michael D. Graham

2024-05-17

Abstract:The dynamics of flexible filaments entrained in flow, important for understanding many biological and industrial processes, are computationally expensive to model with full-physics simulations. This work describes a data-driven technique to create high-fidelity low-dimensional models of flexible fiber dynamics using machine learning; the technique is applied to sedimentation in a quiescent, viscous Newtonian fluid, using results from detailed simulations as the data set. The approach combines an autoencoder neural network architecture to learn a low-dimensional latent representation of the filament shape, with a neural ODE that learns the evolution of the particle in the latent state. The model was designed to model filaments of varying flexibility, characterized by an elasto-gravitational number $\mathcal{B}$, and was trained on a data set containing the evolution of fibers beginning at set angles of inclination. For the range of $\mathcal{B}$ considered here (100-10000), the filament shape dynamics can be represented with high accuracy with only four degrees of freedom, in contrast to the 93 present in the original bead-spring model used to generate the dynamic trajectories. We predict the evolution of fibers set at arbitrary angles and demonstrate that our data-driven model can accurately forecast the evolution of a fiber at both trained and untrained elasto-gravitational numbers.

Fluid Dynamics,Machine Learning
Measuring accurately liquid and tissue surface tension with a compression plate tensiometer

Abbas Mgharbel,Hélène Delanoë-Ayari,Jean-Paul Rieu

DOI: https://doi.org/10.48550/arXiv.0901.4843

2009-01-30

Abstract:Apparent tissue surface tension allows the quantification of cell-cell cohesion and was reported to be a powerful indicator for the cellular rearrangements that take place during embryonic development or for cancer progression. The measurement is realized with a parallel compression plate tensiometer using the capillary laws. Although it was introduced more than a decade ago, it is based on various geometrical or physical approximations. Surprisingly, these approximations have never been tested. Using a novel tensiometer, we compare the two currently used methods to measure tissue surface tension and propose a third one, based on a local polynomial fit (LPF) of the profile of compressed droplets or cell aggregates. We show the importance of measuring the contact angle between the plate and the drop/aggregate to obtain real accurate measurement of surface tension when applying existing methods. We can suspect that many reported values of surface tension are greatly affected because of not handling this parameter properly. We show then the benefit of using the newly introduced LPF method, which is not dependent on this parameter. These findings are confirmed by generating numerically compressed droplet profiles and testing the robustness and the sensitivity to errors of the different methods.

Biological Physics
Drop collision analysis by using many-body dissipative particle dynamics and machine learning

Kaixuan Zhang,Wei Fang,Sang Ye,Zhiyuan Yu,Shuo Chen,Cunjing Lv,Xi-Qiao Feng

DOI: https://doi.org/10.1063/5.0171217

IF: 4

2023-11-13

Applied Physics Letters

Abstract:Droplet collisions are widely observed in daily life and industries. The study of droplet collision dynamics can guide engineering applications in, for examples, inkjet printing, fan cooling, and engine spraying. In this Letter, a numerical simulation method of droplet collision is proposed on the basis of the many-body dissipative particle dynamics. For the collision of two droplets of the same size, the post-collision morphology is analyzed in terms of two key factors: the initial eccentricity parameter and the Weber number. Then, the collision morphology is learned and classified in conjunction with a multilayer perceptron in order to quickly predict the collision morphology from the initial conditions. A machine learning model linking the initial conditions of collision with the post-collision droplet morphology is developed based on three typical morphologies generated by the collision of identical volume droplets. This study provides more insights into droplet dynamics and may benefit related engineering applications.

physics, applied
Physics-Enhanced Multi-fidelity Learning for Optical Surface Imprint

Yongchao Chen

2024-03-22

Abstract:Human fingerprints serve as one unique and powerful characteristic for each person, from which policemen can recognize the identity. Similar to humans, many natural bodies and intrinsic mechanical qualities can also be uniquely identified from surface characteristics. To measure the elasto-plastic properties of one material, one formally sharp indenter is pushed into the measured body under constant force and retracted, leaving a unique residual imprint of the minute size from several micrometers to nanometers. However, one great challenge is how to map the optical image of this residual imprint into the real wanted mechanical properties, \ie, the tensile force curve. In this paper, we propose a novel method to use multi-fidelity neural networks (MFNN) to solve this inverse problem. We first build up the NN model via pure simulation data, and then bridge the sim-to-real gap via transfer learning. Considering the difficulty of collecting real experimental data, we use NN to dig out the unknown physics and also implant the known physics into the transfer learning framework, thus highly improving the model stability and decreasing the data requirement. The final constructed model only needs three-shot calibration of real materials. We tested the final model across 20 real materials and achieved satisfying accuracy. This work serves as one great example of applying machine learning into scientific research, especially under the constraints of data limitation and fidelity variance.

Machine Learning,Artificial Intelligence,Computer Vision and Pattern Recognition,Applied Physics

Quantum interference in the control of molecular processes

Measure liquid surface tension by fitting lying droplet profile

Droplet deformation in steady fluidic flows enables robust, accessible, high-throughput surface tension measurements

Droplet Evolution Prediction in Material Jetting Via Tensor Time Series Analysis

Pinning Effects of Wettability Contrast on Pendant Drops on Chemically Patterned Surfaces

Unsupervised Learning for the Droplet Evolution Prediction and Process Dynamics Understanding in Inkjet Printing

A deep learning perspective on electro-hydrodynamic micro-droplet interface deformation characteristics

Accurate and robust static hydrophobic contact angle measurements using machine learning

Machine learning (ML)-assisted surface tension and oscillation-induced elastic modulus studies of oxide-coated liquid metal (LM) alloys

Design and implementation of an experimental device by the pendant drop method for measuring surface tension of pulmonary fluid

Full-profile Fit Pendent Drop Method for Surface Tension Measurements

Axisymmetric Drop Shape Analysis: Computational Methods for the Measurement of Interfacial Properties from the Shape and Dimensions of Pendant and Sessile Drops.

Deep Learning Method for Determining the Surface Elastic Moduli of Microstructured Solids

Exploring droplet oscillation dynamics in surface tension measurements

Surface tensiometry of phase separated protein and polymer droplets by the sessile drop method

Machine learning materials physics: Deep neural networks trained on elastic free energy data from martensitic microstructures predict homogenized stress fields with high accuracy

Deep Learning Regressors of Surface Properties from Atomic Force Microscopy Nanoindentations

Data-driven low-dimensional model of a sedimenting flexible fiber

Measuring accurately liquid and tissue surface tension with a compression plate tensiometer

Drop collision analysis by using many-body dissipative particle dynamics and machine learning

Physics-Enhanced Multi-fidelity Learning for Optical Surface Imprint